Correlation » Strengths and Weaknesses of Correlations

What you'll learn this session

Study time: 30 minutes

What correlations are and how they're used in psychology
The different types of correlations
The key strengths of correlational research
The important limitations of correlational studies
How to evaluate correlational research in psychology
Real-world applications of correlational studies

Understanding Correlations in Psychology

Correlations are one of the most common research methods used in psychology. They help us understand relationships between variables without directly manipulating them. But like all research methods, they have their strengths and weaknesses that we need to understand.

Key Definitions:

Correlation: A statistical relationship between two variables where a change in one variable is associated with a change in another.
Correlation coefficient: A numerical value between -1 and +1 that indicates the strength and direction of a relationship.
Variables: The factors being measured in a correlational study.

Types of Correlations

🔼 Negative Correlation

When one variable increases, the other decreases. The correlation coefficient will be between 0 and -1.

Example: Hours spent playing video games and exam grades. As gaming time increases, grades might decrease.

🔺 Positive Correlation

When one variable increases, the other also increases. The correlation coefficient will be between 0 and +1.

Example: Hours spent studying and exam grades. As study time increases, grades tend to increase.

Strength of Correlations

The closer the correlation coefficient is to +1 or -1, the stronger the relationship. Values close to 0 indicate weak or no relationship.

👍 Strong

0.7 to 1.0 (or -0.7 to -1.0)

Clear pattern with few outliers

🤝 Moderate

0.3 to 0.7 (or -0.3 to -0.7)

Visible pattern with some outliers

👎 Weak

0 to 0.3 (or 0 to -0.3)

Unclear pattern with many outliers

Strengths of Correlational Research

1. Study of Variables That Cannot Be Manipulated

Correlations allow psychologists to study relationships between variables that would be impossible, impractical, or unethical to manipulate experimentally.

Real-World Example

Researchers want to study the relationship between childhood trauma and adult mental health. It would be unethical to deliberately expose children to trauma, so a correlational study is used instead to examine natural occurrences.

2. Practical Applications

Correlational research has many real-world applications, especially for prediction and assessment.

📊 Prediction

Even though correlation doesn't prove causation, it can still be useful for making predictions. For example, universities use GCSE grades to predict university performance because there's a positive correlation between them.

🚀 Initial Research

Correlations are often used as a starting point for research. If a correlation is found, researchers might then design experiments to test for causation.

3. High Ecological Validity

Correlational studies often use real-world data rather than artificial laboratory settings, which means they can have higher ecological validity than experiments.

For example, studying the correlation between social media use and teenage anxiety using real usage data and self-reported anxiety levels captures genuine behaviour.

4. Collecting Large Amounts of Data

Correlational studies can gather data from large numbers of participants relatively quickly and cheaply, often using surveys or existing datasets.

This allows for more representative samples and more reliable statistical analysis.

Weaknesses of Correlational Research

1. Cannot Establish Causation

The biggest limitation of correlational research is that it cannot prove that one variable causes changes in another.

The Causation Problem

If we find a correlation between ice cream sales and drowning deaths (both increase in summer), we cannot conclude that ice cream causes drowning or vice versa. Both are likely caused by a third variable: hot weather.

💡 Variable A causes B

One possibility

💡 Variable B causes A

Another possibility

💡 Third variable C causes both

Yet another possibility

2. Third Variable Problem

Correlational studies cannot control for all possible variables that might influence the relationship being studied.

For example, a correlation between playing violent video games and aggressive behaviour might be influenced by other factors like family environment or personality traits.

3. Bidirectional Ambiguity

Even if there is a causal relationship, correlational studies cannot determine which variable is the cause and which is the effect.

💫 Example: Depression and Exercise

If we find a negative correlation between exercise and depression (more exercise = less depression), we don't know if:

Exercise reduces depression
Depression reduces motivation to exercise
Both are true (a bidirectional relationship)

4. Sampling Issues

Correlational studies often rely on self-selected samples or convenience samples, which may not be representative of the wider population.

This can limit the generalisability of findings and introduce bias into the results.

Evaluating Correlational Research

When to Use Correlations

Despite their limitations, correlational studies are valuable in psychology when:

Experimental manipulation would be unethical
You want to study naturally occurring relationships
You're in the early stages of research and looking for patterns
You need to make predictions based on relationships

Case Study Focus: Smoking and Lung Cancer

One of the most famous examples of correlational research is the link between smoking and lung cancer. In the 1950s, researchers found a strong positive correlation between cigarette smoking and lung cancer rates. Although this correlation alone couldn't prove causation, it led to further research that eventually established the causal link. This shows how correlational research can be the first step in identifying important health relationships.

Improving Correlational Research

Psychologists can strengthen correlational studies by:

Using large, representative samples
Controlling for potential third variables
Using longitudinal designs (studying the same participants over time)
Combining correlational findings with experimental evidence
Being cautious about causal claims

Summary: Strengths vs Weaknesses

✅ Strengths

Can study variables that cannot be manipulated
Useful for making predictions
Often has high ecological validity
Can collect large amounts of data efficiently
Good starting point for further research

❌ Weaknesses

Cannot establish causation
Vulnerable to the third variable problem
Cannot determine direction of influence
May have sampling issues
Can lead to misinterpretation if not carefully reported

Exam Tip

When evaluating correlational studies in your exam:

Always mention that correlation does not equal causation
Consider alternative explanations for the relationship
Discuss both strengths and limitations
Use specific examples to illustrate your points

🧠 Test Your Knowledge!